Yann N.

hi @Juuustin , Do you know the focal length of the device that took the video ? Yann

Hi @cojosef96 ! Thank you for your PR, it's great to try to speed-up the OpenSfM pipeline by introducing GPU image processing. This has been requested quite sometimes by people...

Hi @cojosef96 , I reworked a bit your branch here : https://github.com/mapillary/OpenSfM/compare/master...feat-polish-sift-gpu More specifically, I refactored a few things : - Unified the Root-SIFT normalisation taking best-of-both-worlds (especially the zero-norm...

That's great, thanks !

Hi @Saijin-Naib I managed to get a nice reconstruction by tweaking the `camera_models.json` `focal_ratio` and set it to `0.36` instead of `0.16`. And using `fisheye` obviously. The value of `0.16`...

Hi @Saijin-Naib , You can add it to the camera database here : https://github.com/mapillary/OpenSfM/commit/9268047c9763a58a4985b51a01b516c8823591db#diff-34d9ae480ee274b0f04757a8f293c760d7cc696d3da333608a92e7ce8cc976b6 Yann

Hi @tlalexander , Can you give a try with the current master https://github.com/mapillary/OpenSfM/tree/357fa7ca3d281f1d07ae29e4899580096023a254 ? Yann

Hi @oliver-batchelor, Unfortunately, OpenSfM only take image position priors as lat/lon pairs, so you would have to fake lat/lon data. Let me know if that helps. Yann

Hi @lmz-whu, Yeah, the OpenSfM team would be more than happy to welcome more feature extracters and matchers. You're welcome to submit a PR to add them. Code-wise, it would...

Hi @Alotaibiuop, Yes, `"projection_type": "spherical"` is the way to go for spherical images. Regarding split and merge, you can use the `large` tools as in the doc : https://opensfm.org/docs/large.html It...